EMPLOYING COMPUTATIONAL LINGUISTIC TECHNOLOGIES AND OCULOGRAPHY TO DEVELOP DIAGNOSTIC TOOL FOR DETECTING...

Area of Science:

• Psycholinguistics
• Computational Linguistics
• Forensic Psychology

Background:

• Early identification of autoaggressive tendencies in young people is critical for suicide risk reduction.
• Psycholinguistic methods, including corpus analysis and eye-tracking, offer promising avenues for diagnostic screening.
• Corpus research identifies generalized speech patterns, while oculography examines perceptual cues linked to suicidal ideation.

Purpose of the Study:

• To develop an algorithmic framework for identifying autoaggressive tendencies in youth using multimodal stimuli.
• To integrate corpus linguistics and eye-tracking data for objective diagnostic markers.
• To enhance early diagnostic screening and suicide prevention programs for at-risk youth.

Main Methods:

• Constructed verbal, visual, and multimodal stimuli based on the idiolect paradigm and forensic authorship attribution.
• Analyzed extensive corpus data (over 100 million tokens) using Python libraries (NLTK, SpaCy) for linguistic markers.
• Integrated stimuli into an eye-tracking application, quantifying gaze delay differences to identify diagnostic relevance.

Main Results:

• Multimodal stimuli and visual cues (characters, portraits) showed significant gaze delay differences in the target group (26-36%).
• Verbal stimuli revealed prolonged gaze on self-referential pronouns (12-25%) and metaphorical death terms, with avoidance of direct suicide terms.
• Developed an automated diagnostic model with weighted coefficients, achieving 72% accuracy in identifying autoaggression.

Conclusions:

• A novel methodology integrating visual, linguistic, and multimodal stimuli with oculography provides objective markers for diagnosing mental health conditions and psychopathological phenomena.
• Oculographic detection of eye movement patterns can identify at-risk states like autoaggression and inform evidence-based diagnostics.
• This approach holds potential for improving suicide prevention strategies, especially for vulnerable young populations exhibiting self-aggressive tendencies.